文章基本信息

标题：A Methodology for Biplots Based on Bootstrapping with R
本地全文：下载
作者：Ana B. Nieto ; M. Purificación Galindo ; Víctor Leiva 等
期刊名称：Revista Colombiana de Estadística
印刷版ISSN：2389-8976
出版年度：2014
卷号：37
期号：2Spe
页码：367-397
DOI：10.15446/rce.v37n2spe.47944
语种：English
出版社：Universidad Nacional de Colombia, sede Bogotá
摘要：A biplot is a graphical representation of two-mode multivariate databased on markers for rows and columns often provided in a two-dimensionalspace. These markers define parameters that help to interpret goodness offit, quality of the representation and variability and relationships betweenvariables. However, such parameters are estimated as point values by thebiplot, thus no information on the accuracy of the corresponding estimatorsis obtained. We propose a graphical methodology, that may be considered asan inferential version of a biplot, based on bootstrap confidence intervals forthe mentioned parameters. We implement our methodology in an R packageand validate it with simulated and real-world data.
关键词：Estadística;Bootstrap Confidence Interval; Graphical Methods; Multivariate Data; Quantiles; Software